Novel Microelectromechanical Systems Image Reversal Fabrication Process Based on Robust SU-8 Masking Layers Scott A

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by epublications@Marquette

Marquette University e-Publications@Marquette Electrical and Computer Engineering Faculty Electrical and Computer Engineering, Department Research and Publications of

7-1-2011 Novel Microelectromechanical Systems Image Reversal Fabrication Process Based on Robust SU-8 Masking Layers Scott A. Ostrow II United State Air Force

Ronald A. Coutu Jr. Marquette University, [email protected]

Published version. Journal of Micro/Nanolithography, MEMS, and MOEMS, Vol. 10, No. 3 (1 July 2011), 033016. DOI. © 2011 SPIE. Used with permission. Ronald A. Coutu, Jr. was affiliated with Air Force Institute of Technology, Wright-Patterson AFB at the time of publication. J. Micro/Nanolith. MEMS MOEMS 10(3), 033016 (Jul–Sep 2011)

Novel microelectromechanical systems image reversal fabrication process based on robust SU-8 masking layers

Scott A. Ostrow II Abstract. This paper discusses a novel fabrication process that uses Ronald A. Coutu, Jr. a combination of negative and positive photoresists with positive tone Air Force Institute of Technology photomasks, resulting in masking layers suitable for bulk micromachining 2950 Hobson Way high-aspect ratio microelectromechanical systems (MEMS) devices. Mi- Wright-Patterson AFB, Ohio 45433 croChem’s negative photoresist NanoTM SU-8 and Clariant’s image rever- E-mail: ronald.coutu@aﬁt.edu sal photoresist AZ 5214E are utilized, along with a barrier layer, to effectively convert a positive photomask into a negative image. This technique utilizes standard photolithography chemicals, equipment, and processes, and opens the door for creating complementary MEMS structures without added fabrication delay and cost. Furthermore, the SU-8 masking layer is robust enough to withstand aggressive etch chemistries needed for fabrication research and development, bulk micromachining high-aspect ratio MEMS structures in silicon substrates, etc. This processing technique was successfully demonstrated by translating a positive photomask to an SU-8 layer that was then utilized as an etching mask for a series of trenches that were micromachined into a silicon substrate. In addition, whereas the SU-8 mask would normally be left in place after processing, a technique utilizing Rohm and Haas MicropositTM S1818 as a release layer has been developed so that the SU-8 masking material can be removed post-etching. C 2011 Society of Photo-Optical Instrumentation Engineers (SPIE). [DOI: 10.1117/1.3625633]

Subject terms: microelectromechanical systems; bulk micromachining; lithography; image reversal; unexposed SU-8; uncrosslinked SU-8. Paper 11017PR received Feb. 26, 2011; revised manuscript received Jul. 14, 2011; accepted for publication Jul. 27, 2011; published online Sep. 2, 2011.

1 Introduction profiles needed for bulk micromaching, such as SF6, when Photolithography is an iterative process used to transfer pat- isotropic silicon etching is required. This paper discusses a terns from a photomask design onto a photosensitive ma- novel processing technique that uses a combination of neg- terial or photoresist. Once transferred into the photoresist, ative and positive photoresists for use with positive masks, the pattern is then developed and “windows” to the underly- resulting in masking layers suitable for bulk micromaching. ing material are opened up. The underlying materials are then The novel process developed here utilizes image reversal etched away, leaving behind a permanent pattern in the lower (IR) and a positive tone mask to achieve the intended re- material, or materials are deposited into the windows. When sults. Of course, one could purchase a new mask if funding fabricating surface micromachined microelectromechanical were available. Often times, however, in university-based, systems (MEMS) devices, thin film depositions are interlaced sponsor-funded research, funding is simply not available to with photolithography, etching, and lift-off processing steps. purchase new masks while device fabrication is on-going. Typically, positive photomasks and photoresists are used in With this novel procedure, the fabrication schedule can be MEMS fabrication to ensure fine resolution and precise min- preserved and research dollars saved. For example, a new μ imum feature sizes. Sometimes, however, reverse field or negative tone mask with 2 m minimum features can cost complementary mechanical structures must be etched into approximately $3500 per mask and takes approximately 1 the substrate (e.g., thermal isolation), which necessitates us- week to write, fabricate, and ship. This unnecessary delay ing a combination of surface and bulk micromachining and and cost can add up quickly when device fabrication-oriented involving aggressive etch chemistries, negative photoresist, research is being conducted. The innovative procedure devel- and robust masking materials. When these situations arise, oped here utilizes photolithography chemicals, equipment, the ability to use a positive photomask with a negative resist and processes that are readily available in most university is often helpful to avoid fabrication delays and higher costs. fabrication facilities. This processing technique is illustrated In addition, certain negative photoresists (i.e., MicroChem’s in Fig. 1, and described in detail in Sec. 3. Specifically, SU-8 NanoTM SU-8) are desirable because when they are hard and Clariant’s image reversal photoresist AZ 5214E are used, baked they become chemically and thermally resistant. This along with a barrier layer composed of a protective positive TM allows them to stand up very well to the aggressive etching photoresist layer (Rohm and Haas Microposit S1818) and an evaporated metal layer, to effectively convert a positive photomask into a negative photomask. The overall process 1932-5150/2011/$25.00 C 2011 SPIE resolution is limited by the SU-8 negative photoresist since

J. Micro/Nanolith. MEMS MOEMS 033016-1 Jul–Sep 2011/Vol. 10(3)

Downloaded From: https://www.spiedigitallibrary.org/journals/Journal-of-Micro/Nanolithography,-MEMS,-and-MOEMS on 11/27/2018 Terms of Use: https://www.spiedigitallibrary.org/terms-of-use Ostrow and Coutu: Novel microelectromechanical systems image reversal fabrication process. . .

to stand up to the aggressive bulk micromachining etching profiles.1 SU-8 is a thick, epoxy-based, high contrast photoresist that is typically used in applications where it will be a permanent layer. Through an exposure and post-exposure bake steps, an SU-8 layer crosslinks and becomes resistant to liq- uid developers, and a wide range of other removal methods [e.g., O2 plasma ashing, reactive ion etching (RIE), corrosive etches, etc.], thus making SU-8 an excellent masking material for bulk micromachining. The exposure step creates an acid and the post-exposure bake step follows this up by thermally activating the acid to crosslink the exposed areas.2 The challenge here is that the SU-8 must remain uncrosslinked, and therefore unexposed, through a majority of the process. If the SU-8 gets exposed at any point it will crosslink and no longer be usable for this process. Various methods have been used to protect an uncrosslinked SU-8 layer during subsequent lithography and/or metal deposition steps, such as a filament evaporated metal layer,3 using an antireflective coating on top of the SU-8,4, 5 UV exposure dose control,6 contact printing of a metal layer,7 and using a positive photoresist as a protection layer.8 These approaches have been used in applications where SU-8 is utilized as a sacrificial material,3 in the creation of microfluidic channels and other stacked structures,6, 9–11 and for electroforming.4, 5 In order to protect the unexposed SU-8 in the processing technique presented here, a protective layer of positive photoresist is utilized on top of the SU-8 to protect it from Fig. 1 Process flow for using positive photomasks to pattern SU-8 being exposed and crosslinked during the evaporation of the masking layers for fabricating inverse MEMS structures. metal barrier layer. This metal layer then serves to protect the unexposed SU-8 from being inadvertently exposed and crosslinked during subsequent UV photolithography steps AZ 5214E allows for feature resolutions below 0.5 μm and performed on the AZ 5214E photoresist layer. SU-8’s resolution limit is approximately 2 μm. Nevertheless, the novel fabrication process presented here is straightforward and suitable for novice device fabricators making bulk 3 Methodology/Procedures micromachined structures. The process developed in this research is illustrated in Fig. 1 above. The process starts with the coating of a clean silicon wafer with SU-8 at the standard spin speeds. A 2 Background 5-μm thick SU-8 layer was utilized in the development of AZ 5214E is a positive photoresist that also has the ability to this novel process. This is followed by a ramped softbake, be utilized for image reversal, providing the negative pattern with a bake at 65 ◦C for 3 min followed by a bake at 110 ◦C of a photomask. It is most commonly utilized in the IR mode for 10 min. This is a longer bake time and higher bake tem- for lift-off processes where thin film metals are deposited and perature than typically prescribed, but it is critical to ensure selectively patterned using a lift-off technique. Through an the integrity of the SU-8 layer and to optimize material com- initial exposure, post-exposure bake (which acts as an image patibility throughout subsequent processing. After a rest of reversal bake), and flood exposure steps, image reversal is several minutes at ambient temperature to allow for the SU-8 easily obtained. After a layer of AZ 5214E is exposed with layer to stabilize, the wafer sample is coated with a layer a positive photomask, it is baked at a temperature between of S1818 positive photoresist, which serves to protect the 115 ◦C and 125 ◦C, which initiates an agent in the photore- SU-8 from being exposed and crosslinked during metal de- sist that crosslinks the areas that have been exposed. This position. Allowing the SU-8 layer to stabilize creates a solid step is critical, since if it is not baked at the right tempera- base for the S1818 layer, and reduces the chances of un- ture ( ± 1 ◦C) the negative pattern of the photomask will not wanted interactions between the two photoresists. The SU-8 be obtained and the AZ 5214E will just act like a positive layer needs to be protected because it is highly sensitive to photoresist. Therefore, it is important to determine and op- UV wavelengths, and high energy photon radiation emitted timize the reversal bake temperature in the range mentioned during the deposition process are in this range and above, above for individual processes. The exposure of a photoac- thus resulting in the exposure and crosslinking of the SU-8 tive compound within the photoresist, and the crosslinking in layer.3 The S1818 layer undergoes a longer softbake than is ◦ 1 the exposed areas, makes these exposed areas insoluble in the usually utilized, 110 C for approximately 12 2 min rather developer. Meanwhile, after a flood exposure, the unexposed than the prescribed 75 s at 110 ◦C. The reason for this longer areas develop away in a standard positive photoresist devel- bake is to create a more stable base for the metal layer. Once oper. Unfortunately, AZ 5214E alone is not robust enough again, the softbake times and temperatures were fine-tuned

J. Micro/Nanolith. MEMS MOEMS 033016-2 Jul–Sep 2011/Vol. 10(3)

Fig. 2 Qualitative difference in integrity of an aluminum barrier layer based on S1818 softbake time. (a) Standard 75 s, shows deforma- tions in the metal layer as a result of solvents baking out from the S1818 layer below it, (b) 12.5 min, shows a more uniform metal layer since the solvents were completely baked out prior to metal deposition.

to optimize material compatibility and the overall process. When a standard S1818 softbake is used, remaining solvents in the photoresist are baked off in subsequent steps following metal deposition, resulting in defects and nonuniform characteristics in the metal layer, as shown in Fig. 2(a) with an 1 aluminum layer. It was discovered that a 12 2 min S1818 Fig. 3 Developed AZ 5214E features on titanium layer. softbake resulted in a more stable and uniform metal layer through subsequent bakes, as shown in Fig. 2(b), while not having a negative impact on the processing of the SU-8 layer. highlights some of the results for the exposure study. As can The sample is then allowed to rest at ambient temperature for be seen, at an exposure time of 60 s, (a), the integrity of the several minutes to allow the S1818 layer to stabilize. Next, layer is poor with multiple defects. At 90 s, (b), the integrity approximately 500 Å of aluminum, titanium, or gold is evap- of the layer is better, but there are still some apparent defects. orated onto the sample to serve as a UV block for the SU-8 At 1 min and 45 s, (c), and 2 min, (d), the integrity of the layer. Aluminum was initially chosen as a metal barrier layer layer is good enough to process with. There is not much dif- because it blocks the 365 nm UV wavelength used to expose ference in the integrity of the layers between 1 min and 45 s the sample, plus it can be deposited via e-beam evaporation and 2 min, so 1 min and 45 s was chosen in order to avoid at a lower power than other available metals, thus reducing any overexposure problems. the likelihood of unintentional crosslinking during the de- The exposure of the photoresist layer is followed by a ◦ position process. Titanium and gold were also investigated ramped post-exposure bake, with a bake at 65 Cfor3min ◦ because they also block the 365 nm UV wavelength, plus followed by a bake at 110 C for 3 min. Once again, this they were shown to effectively not crosslink the SU-8 layer longer bake helps ensure the integrity of the SU-8 layer as it during deposition. proceeds through the process. Next, the remaining AZ 5214E Once the sample has had adequate time to cool, it is coated is removed, as well as the remaining metal and S1818 layers. with AZ 5214E at the standard spin speed for a 1.4 μm layer This opens up the unexposed SU-8 areas for the development and baked at 110 ◦C for 50 s. The sample is then exposed us- step. The sample is placed in SU-8 developer and placed in ing a positive photomask for 3 s at an intensity of 11 mJ/cm2 an ultrasonic bath to develop out the unexposed SU-8 areas, using a Karl Suss MJB3 Photomask Aligner. This is fol- and as a result opening up windows to the substrate, as shown lowed by the most critical step for the AZ 5214E processing, in Fig. 6. The exposed SU-8 areas are crosslinked, so they a 115 ◦C post-exposure bake for 2 min. This post-exposure bake acts as a reversal bake, in which the image reversal characteristic of the AZ 5214E is activated. Following the reversal bake, the sample is subjected to a 14 s ﬂood exposure on a Karl Suss MJB3, which affects the solubility of the AZ 5214E, as described above. The AZ 5214E layer is then developed using 351 Developer. Once the image reversal features are obtained and veriﬁed in the AZ 5214E layer, as shown in Fig. 3, the metal in the open windows is etched with buffered oxide etch (BOE) for aluminum or titanium layers, and gold etchant (Transene Company, type TFA) for the gold layers. The exposed areas of photoresist (S1818 on top of SU-8) are then subjected to a 1 min and 45 s exposure on the MJB3, with the remaining metal features on the sample acting as a UV blocking barrier mask, as shown in Fig. 4.The1min and 45 s exposure time was obtained through an exposure study to determine the optimum time to expose the photoresist layer, since the SU-8 is being exposed through the S1818. This was important because if the SU-8 layer is not exposed completely, and thus not fully crosslinked, it would not have the structural integrity to act as a masking layer. Figure 5 Fig. 4 Titanium barrier layer on photoresist layer.

J. Micro/Nanolith. MEMS MOEMS 033016-3 Jul–Sep 2011/Vol. 10(3)

Fig. 5 Qualitative comparison of SU-8 photoresist layer integrity based on exposure time: (a) 60 s, (b) 90 s, (c) 105 s, and (d) 120 s.

do not develop out, and now are able to act as a mask for the etching. Next, the sample is given a quick BOE dip to clear out any native oxides or contaminants that may have formed in the photoresist windows, and then the features are isotropically etched into the silicon substrate using a RIE tool Fig. 7 Top view of features etched into the silicon substrate utilizing and SF . an SU-8 masking layer fabricated using a positive photomask and a 6 titanium barrier layer. 4 Results and Discussion Figure 6 shows an optical image of the SU-8 masking layer created utilizing this processing technique with a titanium barrier layer. The positive photomask characteristics and features were successfully translated to the SU-8 layer. Figure 7 shows the features obtained with this SU-8 masking layer, trenches etched into a silicon substrate, thus highlight- ing the ability to use a combination of negative and positive photoresists with a positive mask, to create masking layers suitable for bulk micromachining silicon substrates. Figure 8 shows an SEM side view of a trench in the silicon substrate, and Fig. 9 shows an end view of the trench. The SU-8 layer step height was measured at approximately 5 μmpriorto etching. After etching for 60 min, the total step height was measured at approximately 9.2 μm, resulting in an approxi- mate etch depth into the silicon of 8.2 μm. While the ﬁnal process is straightforward, working through the process development showcased some critical Fig. 8 Side view of a trench etched into a silicon substrate (titanium barrier layer used).

Fig. 6 SU-8 masking layer fabricated using a positive photomask and Fig. 9 Endcap of a trench etched into a silicon substrate (titanium titanium barrier layer. barrier layer used).

J. Micro/Nanolith. MEMS MOEMS 033016-4 Jul–Sep 2011/Vol. 10(3)

areas that need to be considered when working with this combination of materials, mostly with how these materials interact with unexposed SU-8. The temperature at which SU- 8 was baked, and the temperatures it was subjected to during processing, were important aspects in developing this process. To begin with, when SU-8 was baked at the prescribed times and temperatures, adhesion issues and cracking were observed during the SU-8 post-exposure bakes. A typical solution to these types of issues is a longer post-exposure bake, but since these issues appeared during the 65 ◦Csoft- bake step, a longer post-exposure bake alone was not a viable option. Through further investigation, it was discovered that extending the softbake and the post-exposure bake times to the steps described above resulted in a consistent and viable SU-8 layer. Another critical area was ensuring that the SU-8 was not inadvertently exposed and crosslinked prematurely. Initially, plans were to use a metal layer directly on top of the unexposed SU-8 as a UV block, therefore only requir- ing three layers—SU-8, metal, and AZ 5214E. Attempts at Fig. 11 Aluminum barrier layer on photoresist layer (both uneven). sputtering and evaporating various metals directly onto the SU-8 resulted in crosslinking. As detailed in Refs. 3, 7, 8, and 10, the sputtering and evaporating processes, through SU-8 provided adequate protection to allow for the evapora- the creation of plasma in the case of sputtering and radiation of a UV barrier metal without exposing and prematurely tion/heat transfer in the case of evaporation, created enough crosslinking the SU-8 layer below. A longer S1818 softbake UV energy to expose and inadvertently crosslink the SU-8. to maximize solvent release and create a more stable base 1 Initial attempts at depositing various protective layers on for the metal barrier layer is crucial. A 12 2 min bake time top of the unexposed SU-8 before depositing the metal layer was found to be a sufﬁcient bake time provided the samples were not successful. Eventually, it was discovered that ex- were not unnecessarily left exposed to ambient conditions. posing both the S1818 and SU-8 layers at the same time, Figure 10 shows the AZ 5214E developed on top of an alu- followed by the SU-8 post-exposure bake, allowed for the minum layer after being inadvertently exposed and not pro- SU-8 layer to be developed out and processed as intended. cessed for approximately 2 days. As can be seen, the alu- Furthermore, when applying the S1818 on top of the SU-8, minum layer develops “hills and valleys,” which translates the spin speed of the S1818 was important. When the S1818 to poor features in the AZ 5214E. This unevenness is seen was spun at a speed greater than what the SU-8 was spun at, through the rest of the process, such as the aluminum mask the S1818 tends to diffuse into the SU-8, resulting in erratic as shown in Fig. 11 and the SU-8 masking layer as shown in thicknesses. When the S1818 was spun at a speed less than Fig. 12. This unevenness results in lower resolution features what the SU-8 was spun at, this issue was overcome and ﬁlm when compared to those that were processed without delays. thickness was as expected. There is most likely, however, This is highlighted in Fig. 13, which shows trenches etched still some material diffusion at the interface of the two pho- into a silicon substrate, and Fig. 14, which shows a side view toresists but it does not negatively affect using the SU-8 and of a trench. Even though features were realized with the S1818 in this process. Therefore, an S1818 layer on top of the

Fig. 12 Uneven SU-8 masking layer fabricated using a positive pho- Fig. 10 Developed AZ 5214E features on uneven aluminum layer. tomask and aluminum barrier layer.

J. Micro/Nanolith. MEMS MOEMS 033016-5 Jul–Sep 2011/Vol. 10(3)

Fig. 15 S1818 release layer for SU-8: (a) developed features in SU- 8; (b) features etched into the silicon substrate; (c) SU-8 removed from sample; (d) proﬁle of etched features in silicon.

sample was cleaved in order to provide a proﬁle of the etched features.

Fig. 13 Features etched into the silicon substrate utilizing an un- 5 Conclusions even SU-8 masking layer fabricated using a positive photomask and A novel processing technique that uses a combination of aluminum barrier layer. negative and positive photoresists with positive tone masks, resulting in reversed masking layers suitable for bulk mi- processing technique utilizing this sample, the results are cromachining, has been developed. This process has been less than desirable. In general, samples that are processed demonstrated through the fabrication of trenches in a sili- the same day provided the best results. con substrate, utilizing a positive photomask along with a The current processing technique results in a robust SU-8 combination of AZ 5214E image reversal photoresist, a bar- layer remaining after etching, which in some applications is rier layer of an S1818 protective layer and evaporated metal not desirable. In order to address this, a quick investigation layer, and SU-8 negative photoresist. Titanium, aluminum, into using S1818 as an SU-8 release layer was initiated, and and gold were shown to be effective UV barrier layer metals. the initial results look promising. In this initial investigation, This novel procedure is actually quite simple and suitable S1818 was spun onto a silicon substrate, followed by an 1 ◦ for even novice device fabricators. Despite being simple and extended bake of 12 2 min at 110 C. The extended bake was straightforward in concept, there are critical pitfalls that must utilized to provide a solid base for the SU-8 layer that is spun be considered and effectively avoided when working through on top of the S1818. After the SU-8 was spun on, the sample this process. These include, but are not limited to, thermal was exposed for 10 s, and then developed out as shown in concerns with the SU-8 layer to ensure it does not hard bake, Fig. 15(a). The features were then etched into the substrate negative and unforeseen interactions between the SU-8 layer using an RIE, as shown in Fig. 15(b). This was followed by and other material layers, and ensuring that the uncrosslinked placing the sample in acetone and placing it in an ultrasonic SU-8 does not get exposed and crosslinked prematurely. Fi- bath. In less than 2 min, the S1818 dissolved and the SU-8 nally, a simple process for removing unwanted robust layers layer floated off of the substrate leaving behind a sample of SU-8 (post-etch) was developed where S1818 was used as with bulk micromachined features without an SU-8 masking releasable layer deposited underneath the SU-8. layer, as shown in Figs. 15(c) and 15(d).InFig.15(d),the Acknowledgments Special thanks to Mr. Paul Cassity for his tireless assistance in working through this development process, and for his constant advice and guidance. His insights, willingness to brainstorm and roll up his sleeves, and positive attitude con- tributed greatly to the success of this research. Thanks also go to Captain Derrick Langley for his insight and guidance on processing techniques, Mr. Rich Johnston and Mr. Rick Patton for their assistance in depositing the metal layers, and Second Lieutenant Jack Lombardi for his assistance in ob- taining SEM images. The views expressed in this paper are those of the authors and do not reflect the official policy or position of the United States Air Force, Department of Defense, or the U.S. Government.

References 1. AZ 5214E Image Reversal Photoresist Product Data Sheet. 2. Nano SU-8 Negative Tone Photoresist Formulations 2–25, Rev. 2/02. http://www.microchem.com/products/pdf/SU8_2–25.pdf. Fig. 14 Side view of a trench etched into a silicon substrate (Al barrier 3. C. Chung and M. Allen, “Uncrosslinked SU-8 as a sacriﬁcial material,” layer used). J. Micromech. Microeng. 15(1), N1–N5 (2005).

J. Micro/Nanolith. MEMS MOEMS 033016-6 Jul–Sep 2011/Vol. 10(3)

4. C. K. Chung, Y. Z. Hong, and W. T. Chang, “Fabrication of the mono- Scott A. Ostrow II is an Air Force Captain currently assigned to lithic polymer-metal microstructure by the backside exposure and elec- the National Security Administration (NSA) at Fort Meade, Maryland. troforming technology,” Microsyst. Technol. 13(5/6), 531–536 (2007). He earned his undergraduate degree in electrical engineering from 5. C. K. Chung, C. J. Lin, L. H. Wu, Y. J. Fang, and Y. Z. Hong, “Se- the University of Portland, Portland, Oregon in 2005 and his Masters lection of mold materials for electroforming of monolithic two-layer in electrical engineering from the Air Force Institute of Technology, microstructure,” Microsyst. Technol. 10(6/7), 467–471 (2004). 6. F. G. Tseng, Y. J. Chuang, and W. K. Lin, “A novel fabrication method Wright-Patterson AFB, Ohio in 2011. His current research interests of embedded micro channels employing simple UV dosage control and are MEMS and microelectronics device fabrication. antireflection coating,” in The Fifteenth IEEE International Conference on MEMS, pp. 69–72 (2002). 7. D. Haefliger and A. Boisen, “Three-dimensional microfabrication in Ronald A. Coutu, Jr. is an assistant profes- negative resist using printed masks,” J. Micromech. Microeng. 16(5), sor of Electrical Engineering and the clean- 951–957 (2006). 8. V. M. B. Carballo, M. Chefdeville, M. Fransen, H. van der Graaf, room director at the Air Force Institute of J. Melai, C. Salm, J. Schmitz, and J. Timmermans, “A radiation imag- Technology (AFIT) at Wright-Patterson Air ing detector made by postprocessing a standard CMOS chip,” IEEE Force Base (AFB), Ohio. He received his BS Electron Device Lett. 29(6), 585–587 (2008). in electrical engineering from the University 9. B. E. J. Alderman, C. M. Mann, D. P. Steenson, and J. M. Chamberlain, of Massachusetts at Amherst in 1993, his MS “Microfabrication of channels using an embedded mask in negative in electrical engineering from the California resist,” J. Micromech. Microeng. 11(6), 703–705 (2001). Polytechnic State University (CalPoly) in San 10. F. Ceyssens and R. Puers, “Creating multi-layered structures with free- Luis Obispo in 1995, and his PhD in elec- standing parts in SU-8,” J. Micromech. Microeng. 16(6), S19–S23 (2006). trical engineering from the AFIT at Wright- 11. P. Abgrall, C. Lattes, V. Coned´ era,´ X. Dollat, S. Colin, and A. M. Gue,´ Patterson AFB in 2004. He is a California registered professional “A novel fabrication method of flexible and monolithic 3D microfluidic engineer in electrical engineering, a senior member of the IEEE and structures using lamination of SU-8 films,” J. Micromech. Microeng. a member of SPIE. His current research interests are MEMS fabri- 16(1), 113–121 (2006). cation and microelectric contacts.

J. Micro/Nanolith. MEMS MOEMS 033016-7 Jul–Sep 2011/Vol. 10(3)

Downloaded From: https://www.spiedigitallibrary.org/journals/Journal-of-Micro/Nanolithography,-MEMS,-and-MOEMS on 11/27/2018 Terms of Use: https://www.spiedigitallibrary.org/terms-of-use